The long-lasting viral reservoir in resting CD4 memory cells and other anatomic sanctuaries requires that antiretroviral therapy (ART) be continued life-long. The recent report of the infant in Mississippi who started treatment within 30 hours of birth and who has been able to maintain viral suppression off treatment raises the tantalizing possibility that establishment of the viral reservoir may be avoidable in some infants. Limitations of this single case report necessitate urgent replication as the potential public healt benefits of the intervention are profound. We propose a single-arm clinical trial in Johannesburg, South Africa, to replicate the Mississippi case in a more robust number of infants. Our trial is designed to test whether initiation of ART within 48 hours of birth has the potential to allow the majority of HIV-infected infants to safely discontinue ART without viral rebound. The context of starting ART within hours of birth and stopping it 18 months later provides an unprecedented opportunity to elucidate mechanisms involved in the establishment and maintenance of viral reservoir. In the context of this unique trial, we will collect samples before, during and after AR cessation to investigate potential mechanisms driving the establishment of the viral reservoir. Following the exciting new findings from the Mississippi child, we hypothesize that infant immunological developmental maturity is one of the critical parameters influencing the success of early treatment. ART given at birth is not only close in time to acquisition of infection but is given during a developmentally-critical time period when the immune system is transitioning to its mature form and is at its most quiescent. We will also recruit a carefully-selected observational cohort of infants 4-12 weeks of age initiating and continuing ART in the standard way from routine clinical services. We will investigate whether lesser CCR5 and CD2 expression on T-cells, a smaller pool of CD4 memory cells, reduced markers of T-cell activation, and larger proportions of regulatory (suppressive) T-cells (T-regs) relative to Th17 cells will be associated with more limited seeding and greater decay of the viral reservoir when ART is started at a young age. We will also investigate whether markers in infant stool samples can be used as a non-invasive method of defining relevant immune and HIV-specific parameters associated with smaller viral reservoirs.